Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 03:55:22 PM

Login with username, password and session length

Recent Posts

 

Preamplifier Input Impedance

Started by tonyharker, February 22, 2014, 05:50:02 PM

Previous topic - Next topic

tonyharker

Guitar preamplifiers often have a very high input impedance >500k but what is the optimum impedance to load a pickup.
Is it higher or lower than this assuming the pickups are not active types.

Edited to remove typo.

Tony

phatt

It's a catch 22 in My opinion and depends on many factors. xP

Input Z Too low and you loose bandwidth and signal swing,, Too high and you make resonant humps extreme and raise the sensitivity and bandwidth and that can send you nuts trying to rectify inside the circuitry.

Roly has kindly gone and done some good work on this pu and preamp interface issue, might give you some idea of what goes on inside. :dbtu:

http://www.ozvalveamps.org/pickups.htm

I would add some considered thought to it all;
In the early days of Valves making the input very sensitive made sense as there was no choice with often only 1 or 2 stages before poweramp.
But these days even cheaper gear often have multiple stages so the issue of max swing is not so relevant.

If anything, really hi Z input + hi gain PU + multipule gain stages + highly efficiant speakers + many pedals. ++++ get the picture? 8|
And still no one seems to be able to find the tone/sound they want. xP

More Does not go on forever and my guess, it may actually get in the way of great tone.
I've found that depending on how well you setup the circuits you design,, from 220k up to 500k may work well for controlled sound. (YMMV)
Phil.

tonyharker

Thanks phatt, that article has really given me something to think about.

:) Tony.

phatt


Well good to hear.
Something as simple as a cable can destroy a good tone. ???

Cesar Diaz once purchased a whole lot of top of the range guitar cables for Stevie Ray Vaughan.
Stevie hated them and refused to use them.
His words where; These new cables pass too much electricity they are horrible and destroy my tone. :grr

The issue was they passed much more hi frequency content which seems to be a large part of the harshness associated with hi gain amplification on guitar Amplifiers.

So that is how easy it is to get tricked up. 8)
Phil.

QReuCk

If Roly checks that topic, I would be very interested in a confirmation that all these diagrams are made without a passive volume and tone circuit in the guitar (straight from pickup to jack). These circuits in guitars will have an impact on these curves so while we are at it, I'll try to explain my thoughts about it (and hope more knowledgeable people than me will either confirm or correct me if I'm wrong).

Regarding the volume pot of your guitar:
Let's assume you have no tone circuit for now or a no load tone pot on 10. With volume on 10, you effectively have the pot value in parallel with the input resistance of the amp. If your volume pot is a 500K, then you have 500K in parallel with whatever input impedance your amp has.
Aside from custom wiring, and other trebble bleed, 2 arrangements are most common:
1° gibson volume pots have the hot from pickup on an outer lug, ground on the ather outer lug and wiper to guitar plug or switch. When the wiper is at half resistance, your pickup is loaded with 250k in series with 250K of the other half of your pot which is in parallel with your amp input. So that's 250K + 1 / (1/250+1/Z)
2° "indonesian wiring": pickup hot to the wiper and guitar plug to one outer lug. When at half resitance, you have 250K in parallel with your amp input which is in series with 250K. So that's
1/(1/250K + 1/(250K+Z)).

Your volume knob doesn't only reduce the amplitude of the signal, it also changes the load of the pickup.

As a side note, I have a guitar without a tone pot. I wired the volume so that it's just a variable bypass to ground (wiper to guitar plug, one outer lug to ground). Then the variable value is in parallel with the amp input and the only thing it does is changing the load. This has a pretty dramatic effect on frequency content in addition to volume (remember electricity loves the path of least resistance), which I happen to like for that guitar because I often need more trebble when the signal needs to be hotter. I think I need to change the pot value though cause the usable tones are more in the lowest half of the range than anywhere else.

Then you enter in the equation what your passive tone circuit does and it's actually another can of worm alltogether. Base line for now is that it also modifies the load in addition to letting a cap bleed some frequencies to ground.

Now back to your concern. By experience, I would say that a too low input impedance will sound lifeless and that a too high one might sound just unmanageable (pretty much what Phatt said really). I would say that combined with the passive volume and tone circuitry, 1M is just fine. That's the value most guitar manufacturers have in mind when selecting what volume and tone pots they put in guitars anyway and we have to assume some of them know their stuff.

Roly

Yes all those plots on "pickups" were for pickup-direct, no hidden controls, and yes they do have an influence (and I'm inclined to the view that most guitar pots are too low value, e.g. 250k volume).

The very short of it I have discovered that Leo Fender was a much smarter cookie that perhaps many give him credit for.  I had a long hard look at his early "harmonic vibrato" and it's full of hidden cunning; then I had a look at Tele pickups and concluded that this upper resonance is what he was playing with on that rig, and out of which the Broad/Tele/Strat sound was born.  He didn't invent the pickup, but I think he found a sonic magic trick.


A Tele-fanatic friend loaned me a Fender history and I eventually got to a table of pickup parameters that for the first time contained inductances that made my eyes pop, totally obvious for a thousand turns, the significance just never struck me before.  With inductances like that they must be resonant in the audio band!  <3) And sure enough a bit of calculation and modeling showed they were significant resonances.

Better, you could play around with them without having to gut your guitar, different shunt R and C.  (and I've never been satisfied by guitar "tone" controls).

While one Meg has been a traditional and reasonable value (despite being strictly speaking off the twin-triode datasheet), I started looking at the effects of using much higher input resistances, e.g. 2M7, 4M7, .

This led by stages (and some being prodded by sundry AGGH forum members, thanks guys!) to The Gronk Box - multi-meg input JFET in to four transistor buffers driving isolation transformers.  Should be as close to the pickups as possible, built in, belt pack, or floor with short low-C lead.


The AVA "pickups" page is not intended as a prescription, but to expose/explore another ignored area where players can easily experiment with their "tonality".

I actually only discovered the Vari-Tone, which is down a similar pathway, by accident very late in my research, but it would seem that Gibson was already where I was going.

Gronk Vari-C;


Gronk Vari-CR;

If you say theory and practice don't agree you haven't applied enough theory.

QReuCk

I think I understand where you are going with this page. That's very usefull indeed as except for it and this forum, input impedances are often totally overlooked.

Regarding the resonant peak: I do believe frequency response curves doesn't tell the whole story. For me the "enveloppe" of the resonance isn't the same as one of a non-resonant band. i.e.: What happens after you pick your string (how the harmonics develop and fade out is a bit different for the ones in the frequency peak region and the ones that are out of it). Anyway, I find use for a high resonant peak sometimes. But to my liking, tone and volume on 10 on a strat bridge pickup (well I wired mine to tone pot) and into a 1M input impedance is about as much of a resonant peak I like, and I even find that a little less volume (so less load) and tone (peak shifted to lower freq and reduced in value) might be desirable with some levels of drive in some songs.
As a side note: with standard tone control, a new peak appears in the bass/low-mid region when tone pot is close to 0. I did not really checked the maths behind it, but this peak is clearly audible if you pay attention when rolling down your tone (Fender even made a special tone circuit named grease bucket to prevent that as this peak is far from sounding right).

I don't know if that makes sense, but that's what I hear. I know for sure the ear is much less than a good measuring instrument for these stuff (psycho accoustics, etc...), but my ears happen to be the medium between what comes out of the speaker and my pleasure, so I still have to trust them in some way, even if they are probably wrong. That sure isn't to say other people should do just like me. Their ears are not mine and they probably play differently (better) than me. So I encourage them to experiment with these things for themselves and decide what they want.

Regarding your Gronk-O-Matik: That's too much of a layout for me to fully understand, but I like the idea.

Roly

Vari-GronkC and Vari-GronkCR.

Conceptually it's an adjustable resistive and capacitive load box which could replace current guitar controls by acting directly on the pickup; a quasi-parametric in effect with C altering the frequency and R altering the amplitude - not so much "EQ" as tonestacking the pickup.

I have certainly found with the Gronk FET buffer/splitter/isolator experiments that my cheap bench test guitar ($100 "Casino" Strat copy) is more lively into 10Meg, but to apply it properly requires it be built in to the guitar, and all that implies.

These work via a process called "bootsrapping" where a signal is applied to the normally ground ends of some components such as resistor and capacitors.  It is based on the idea that any component with identical voltage on each end isn't actually there, becomes an open circuit.

If you feedback 100% of the input the component is open, 50% then only half-there, etc.  By doing this with a resistor and capacitor I get one of each across the input (pickup) and can vary their effective values from open circuit to their actual values, in this case multi-meg leakage-to-R3 100k and strays-to-C4 1nF (=1000pF) as two independent continuous controls (unlike the VariTone).

Still think Don Tillman's idea is a killer. 



http://www.till.com/articles/PreampCable/

The original Gronk-O-Matic


If you say theory and practice don't agree you haven't applied enough theory.